SOLVED Cobalt Drive problems…..SOLVED……..FINALLY!

[Jeff C]

New member here and also new to building pedals, as you will shortly see. I previously built a BYOC Reverb pedal kit successfully and it sounds great. I’ve also built from scratch a Strat, a Tele, a 59 LP, and an LP Junior. So, I wanted a blues driver pedal and thought I’d take a whack at the Cobalt Drive from PedalPCB. The directions are not as detailed as in a kit so I’ve had some difficulty. Clearly in over my head. I managed to procure the parts from the list (not without trial and error on my part) from Tayda. I mistakenly bought a switch without solder lugs so after consulting a patient, helpful friend, bought a daughter board and did my best to figure out how to marry the two - having no skills whatsoever in electronics. History major. I have managed to assemble it but I get nothing from it except static pops when I plug it in. Bypass doesn’t work. No idea what I’ve done wrong but I suspect it revolves around needing to do more with the daughter board on the switch. But what do I know? Here’s a couple of pics (the power jack is not there but I did connect it to test). And I haven’t installed the LED yet. Any and all suggestions are eagerly welcomed. You may recomemd I should just start over but I hope not….. Thank you!

 

[Chuck D. Bones]

JFET-based pedals are not necessarily plug & play because there is too much part-to-part variation within the same part number. The JFETs used in the BD-2 and clones need to be selected and/or resistor values tweaked for best performance. I recommend the JFET pairs should be matched for Vp. Q5-C and Q6-C should be as close to 4.5V as possible. Severely mis-matched JFETs will prevent this circuit from working.

I wrote a fairly lengthy treatise on the BD-2 in Chucks Boneyard. Included in the discussion are some mods and the procedure for balancing the JFETs.

 

[Mcknib]

Damn how'd I miss that

I'm forever telling people to go check Chucks Boneyard.....

And I missed it! coulda saved a few pages here reading a few pages over there

Anyway my complimentarys to you @Chuck D. Bones or is it differentials

 

[Jeff C]

JFET-based pedals are not necessarily plug & play because there is too much part-to-part variation within the same part number. The JFETs used in the BD-2 and clones need to be selected and/or resistor values tweaked for best performance. I recommend the JFET pairs should be matched for Vp. Q5-C and Q6-C should be as close to 4.5V as possible. Severely mis-matched JFETs will prevent this circuit from working.

I wrote a fairly lengthy treatise on the BD-2 in Chucks Boneyard. Included in the discussion are some mods and the procedure for balancing the JFETs.

Whoa. Gonna need to go back to school for that one. Here’s to hoping my new Q5 when it arrives matches Q6…….

 

[Chuck D. Bones]

Your Q6 doesn't even match itself. Every time you measure it you get a different voltage.

Dirty board?
Bad connection?
Interference?
Oscillation?
Demonic possession?

 

[Jeff C]

Your Q6 doesn't even match itself. Every time you measure it you get a different voltage.

Dirty board?
Bad connection?
Interference?
Oscillation?
Demonic possession?

I vote for demonic possession.

 

[Chuck D. Bones]

Whatever is going on with Q6 and the whole 2nd stage, we need to get it fixed first. We don't need Q5 to do that.

First off, from what I see your board is a mess. Remove ALL of the flux residue with alcohol and a toothbrush. Flux absorbs moisture and creates current leakage paths. That could very well be the reason Q6's bias keeps wandering around. Also, you will be able to see more when you inspect the board for debris and bad solder joints. Oversized components also make inspection very difficult.

I gotta say that I don't appreciate having to wade thru memes to get to the actual information. If this thread continues to be a circus then I'll go find something productive to do, like change out the radiator and A/C condenser on my Jag.

 

[Jeff C]

Haven’t posted any memes, Chuck, and I do appreciate your patient help. I have cleaned the board since the photos early in this thread but will do so again, with more gusto.

 

[Chuck D. Bones]

Post some more detailed pix when you do, maybe one of us will spot something you missed.

You probably already said, but where did you get the J201s on the SMD boards?

Yeah, I know it wasn't you. There's always a few jokers in the deck they know who they are.

 

[Jeff C]

Post some more detailed pix when you do, maybe one of us will spot something you missed.

You probably already said, but where did you get the J201s on the SMD boards?

Yeah, I know it wasn't you. There's always a few jokers in the deck they know who they are.

J201’s came from PedalPCB . See #115 above. I previously had some from Amazon which caused folks to suggest I swap them out, so I did.

 

[Chuck D. Bones]

Robert sells good stuff and they come pre-assembled, so that's one less thing to worry about.

 

[Jeff C]

Robert sells good stuff and they come pre-assembled, so that's one less thing to worry about.

Well, I did have to solder on the pins and then put ‘em on the board so there is some amateur work there…..

 

[Jeff C]

Close ups, per your request:

 

[Jeff C]

I checked all the spots. Nothing cracked or loose. Today’s Q6 voltages: 9, 8, 5. Same with gain at 0 or 10

 

[giovanni]

I wonder if that oversized resistor's lead is sometimes touching the adjacent transistor's lead?

 

[Chuck D. Bones]

So now Q6-C is more-or-less correct. We'll see how long that lasts. Are you rounding-off the voltage readings?

I guess I'm not asking for the right thing. 640x480 is too low-res to be useful. What's with the jumper under Q1 & Q3, is that a repair for a broken trace? Any other broken traces or lifted pads? I can't tell from these pix. Pix need to be at least 1600 x 1200 to be useful.

 

[Jeff C]

I wonder if that oversized resistor's lead is sometimes touching the adjacent transistor's lead?

It is not, fortunately.

 

[Jeff C]

So now Q6-C is more-or-less correct. We'll see how long that lasts. Are you rounding-off the voltage readings?

I guess I'm not asking for the right thing. 640x480 is too low-res to be useful. What's with the jumper under Q1 & Q3, is that a repair for a broken trace? Any other broken traces or lifted pads? I can't tell from these pix. Pix need to be at least 1600 x 1200 to be useful.

Yes, I’m rounding off. Sorry, taking pics with the mag lite app but that’s about as sharp as I can get. Per #320 above, I decided to put that bridge in between Q1 and Q3 and amazingly I immediately got audio out to my amp, albeit pretty low volume. And the pots worked. Since then,,,,,,,I have somehow lost audio out to the amp. I’ve looked at the traces and apart from Q1-Q3, they look alright. Pads are all still there, although a couple seem a bit scorched from all my de-soldering and removal activity……

 

[Chuck D. Bones]

I hate to say this, but after all this time and so little progress, I think it's time to punt. Get a new board, the right size parts and start over. The only parts I'd reuse are the J201s and the pots, if you can get them off without damaging them.

 

[Jeff C]

Afraid you would say that. Ah well, hate to admit defeat but I guess you’re right. Much appreciate everyone’s help!

 

[Chuck D. Bones]

Consider this one a practice run for the next build.

My main concern at this point is collateral damage. The board has suffered enough already. If we continue to touch-up solder joints & replace parts, there's a high risk of lifted pads. Sure, they can be repaired, but is it worth the effort of finding and fixing every last one?

If you don't already have a CCTT (Cheap Chinese Transistor Tester), then you would be well advised to get one. Pretty essential when building pedals containing FETs or Germanium transistors. They're around $20 on eBay. At least they were a couple of years ago when the dollar was worth something.

I don't know what kind of phone you have, but if you have a smart phone built in this century it will have a camera capable of much higher resolution than 640x480. Otherwise, there are pocket digital cameras for under $100.

Al lastly, when reporting DMM readings, report every digit. Sometimes it's that last 10mV we're interested in.

Good luck on the next build!